Collaborative Research and Mapping for Nairobi’s Public Transit
Technology is transforming the way we relate to transit. Digital Matatus is a key player in the movement to make transportation in the developing world more efficient, and open. A collaborative project of the University of Nairobi, Columbia University’s Center for Sustainable Urban Development, MIT’s Civic Data Design Lab, and Groupshot, we are collecting and standardizing transit data for Nairobi’s Matatu system—the city’s decentralized city bus system– making it open and available to the public for the first time.
Digital Matatus has developed and deployed purpose-built mobile phone apps to more efficiently collect location data of Nairobi’s buses. We have used the information gathered through this crowd sourcing method to produce standardized bus routes for the city, now assembled together and made available in a full city-wide bus map released to the public in January of 2014. We are continuing our efforts in the city to refine, expand, improve, and achieve self-sustainability of our data collection process to be able to best serve riders and drivers. The next step is to expand our tools and methodology to the global scale.
Digital Matatus has taken advantage of the dramatic increase in the power and functionality of mobile phone technology. Using highly affordable, off-the-shelf mobile phone apps, namely Google’s MyTracks and Conveyal’s TransitWand, our team members at the University of Nairobi have successfully captured route data for the city’s Matatu bus lines. This data was then translated into Google’s General Transit Feed Specification (GTFS) standard and used to create a full city-wide map. The Digital Matatus project is the first such comprehensive mapping of Nairobi’s bus network, as well as the first integration of paratransit data into the GTFS standard.
The first step in this work involved compiling existing data on Matatu routes. After querying all of the relevant agencies, including the Transport Licensing Board, it became clear that the existing government and agency data was incomplete, outdated, and often inaccurate—especially given the large number of road projects and new real estate developments in and around the city. These were prompting the Matatu industry to alter existing routes or devise new ones constantly. Additionally, what data did exist employed inconsistent methodologies and data structures that made the data rather impossible to combine.
Once we realized that the existing data was not consistent, reliable, or comprehensive, it became clear that the bus and Matatu route data for the city would need to be collected wholly from scratch. We devised a standard protocol and clear methodology for creating a route map along with a GTFS compatible data structure. In all cases, student team members from the University of Nairobi would ride a Matatu route, mobile device in hand, and would collect data using one of the pre-selected collecting apps. These apps were used to generate latitude and longitude points along the route, and record the location of stops along with specific meta-data about each stop. While many paratransit systems involve stopping at varied locations depending on customer demand, regular and central stops at large terminals exist within this system in Nairobi. Students identified a stop based on their personal knowledge, information from frequent users of these routes, visual notation (signs, shelters etc.), and if necessary, confirmation from discussion with an operator or group of commuters on a route.
Across Nairobi, Matatu routes were organized into eight major corridors used to codify clear routes and bus stops. We used this structure to develop a coding system based on branching and rotation to give each stop and route a unique, logical identifier. This was important as previously, stops were not identifiable in a uniform way, with even stops shared by several routes identified separately. This form of coding was an important step in adding a level of standardization to the system that could significantly improve user interaction.
Once the data was collected, it need to be cleaned and formatted into GTFS. However, several typically required data points for the standard simply do not exist for the Matatu system—data such as calendars, service frequencies and operating schedules. In addition, fares are variable, and routes and stops can change depending on traffic patterns, police checks, commuter demands or the prevailing weather – for instance, when it rains, fares can triple and routes are adjusted to avoid traffic jams. Working with Google, we have encouraged the company to make the GTFS standard more flexible, allowing for systems such as Nairobi’s to be more easily included.
To date, Digital Matatus has collected data for 120 matatu routes as well as 6 formal rail routes, with which we have produced Nairobi’s first city-wide transit map. In addition, we have shared our transit data over the publicly accessible GTFS exchange.
The mission of the Center for Sustainable Urban Development (CSUD) is to engage in and foster education and research for the advancement of physically and socially sustainable cities. CSUD concentrates on understanding, articulating and disseminating information about the unique and important challenges and opportunities of urban development in low- and middle-income countries, as well as in its home community of Metropolitan New York.
The Civic Data Design Lab works with data to understand it for Public Good. We seek to develop alternative practices which can make the work we do with data and images richer, smarter, more relevant, and more responsive to the needs and interested of citizens traditionally on the margins of policy development. In this practice we experiment with and develop data visualization and collection tools that allow us to highlight urban phenomenon. Our methods borrow from the traditions of science and design by using spatial analytics to expose patterns and communicating those results, through design, to new audiences. The Civic Data Design Lab is a based in the Urban Planning Department of the Massachusetts Institute of Technology (MIT).
The School of Computing and Informatics (SCI) is part of the College of Biological and Physical Sciences of the University of Nairobi. The School of Computing and Informatics seeks to be a leading centre of excellence in research, research and development (R&D) and advanced education in computing; addressing national and regional needs for information and communications technology and product development. The School has several ongoing partnerships and collaborations working on various research and development projects.
Groupshot is an innovative project design and development firm based in Cambridge, MA. Groupshot creates, researches, and develops technology-driven projects that interface with and support existing local systems. The firm has worked extensively in Nairobi on innovative technology projects and informal systems in collaboration with groups such as Ushahidi, FrontlineSMS, the NaiLab and the iHub. Groupshot is also a World Bank Innovation winner for their development-oriented technology ideas. Groupshot works with a range of partners and organizations across Nairobi and has lead projects and research relating to data, transport, mapping, design, and technology literacy. They have also mentored students and professional projects in Nairobi as part of the Institute for Global Leadership at Tufts University.
Jacqueline M. Klopp is an Associate Research Scholar at the Earth Institute, Columbia University. A political scientist with extensive experience in Kenya, she works on the political economy of transportation in Nairobi at the Center for Sustainable Urban Development, a Volvo Education and Research Foundations center of excellence on Sustainable Urban Transport.
Dan Orwa is a lecturer at the School of Computing and Informatics in the University of Nairobi. His PhD research was in ICT Adoption. His research interests are in mobile phone applications, ICT4D, M4D, ICT adoption among rural communities. He is a technical committee member of the International Federation of Information Processing (IFIP) representing Kenya, a publication reviewer with ICT4D and a member of Computer Society of Kenya, a professional member of ACM, a member of VeSeL and a member of the C4D Lab
Peter Waiganjo Wagacha is an Associate Professor at the School of Computing and Informatics, University of Nairobi. He has worked on a few transport-related student research projects. His research interests include data mining, ICT4D research, M4D and data-driven human language technology and mobile technology. He has also conducted numerous mobile programming bootcamps for students and is a founder member of the C4D Lab at his School.
Sarah Williams is currently an Assistant Professor of Urban Planning and the Director of the Civic Data Design Project at Massachusetts Institute of Technology’s (MIT) School of Architecture and Planning School. The Civic Data Design Project employs data visualization and mapping techniques to expose and communicate urban patterns and policy issues to broader audiences. Before coming to MIT Williams was Co-Director of the Spatial Information Design Lab at Columbia University’s. Sarah has won numerous awards including being named top 25 urban planners working in technology and 2012 Game Changer by Metropolis Magazine. Her work is currently on view in the Museum of Modern Art (MoMA), New York.
Adam White, co-founder of Groupshot, has worked on and consulted a variety of innovative development, infrastructure, and urban focused projects in over half a dozen countries including Kenya. He completed his Master’s Degree in City Design and Social Science at the London School of Economics. He graduated with honors from the Tufts School of Engineering with a focus in Development and Technology. He has worked extensively in Nairobi.